INDEX

Explanations

dialogue and interactions between characters, particularly involving emotions and relationships

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ombok

-0.14

çħ§

-0.14

 kÃ¥

-0.14

ysl

-0.13

ÑĭÐ²Ð°

-0.13

¶Į

-0.13

spb

-0.13

ØŃØ©

-0.13

Performed

-0.13

isko

-0.13

POSITIVE LOGITS

had

0.85

had

0.78

 hadn

0.78

Had

0.68

 habÃŃa

0.65

Had

0.61

 hade

0.58

 avait

0.54

 hatte

0.53

 hadde

0.52

Activations Density 0.621%

dialogue and interactions between characters, particularly involving emotions and relationships

No Comments

No Known Activations